Model-based segmentation of spatial cylindrical data

ABSTRACT

A new hidden Markov random field model is proposed for the analysis of cylindrical spatial series, i.e. bivariate spatial series of intensities and angles. It allows us to segment cylindrical spatial series according to a finite number of latent classes that represent the conditional distributions of the data under specific environmental conditions. The model parsimoniously accommodates circular–linear correlation, multimodality, skewness and spatial autocorrelation. A numerically tractable expectation–maximization algorithm is provided to compute parameter estimates by exploiting a mean-field approximation of the complete-data log-likelihood function. These methods are illustrated on a case study of marine currents in the Adriatic sea.     

Dependence Structure of Spacings of Order Statistics

Let X_1:n ≤ X_2:n ≤···≤ X_n:n denote the order statistics of a sample of n independent random variables X₁, X₂,…, X_n, all identically distributed as some X. It is shown that if X has a log-convex [log-concave] density function, then the general spacing vector (X_k1:n, X_k2:n ? X_k1:n,…, X_kr:n ? X_kr?1:n) is MTP₂ [S-MRR₂] whenever 1 ≤ k₁ < k₂ <···< k_r ≤ n and 1 ≤ r ≤ n. Multivariate likelihood ratio ordering of such general spacing vectors corresponding to two random samples is also considered. These extend some of the results in the literature for usual spacing vectors.     

Making the Transition from Welfare to Work: Employment Experiences of CalWORKs Participants in the San Francisco Bay Area

ABSTRACT

This study describes welfare-to-work participants in the San Francisco Bay Area, support services, experiences with the CalWORKs program, and predictors of employment status in the wake of welfare reform. Findings indicate that many are working and more Stayers and Recidivists than Leavers are using food stamps and Medi-Cal. Multivariate analysis reveals that race and financial supports were the significant factors contributing to employability, defined as the ability to secure employment despite the need to supplement earned income with welfare payments. To help people stay off of welfare, case management services are needed to help participants maintain employment and increase job skills. In addition to expanding our understanding of human behavior within the social environment of poverty, implications for practice and policy are identified.     

Influence diagnostics on the coefficient of variation of elliptically contoured distributions

In this article, we study the behavior of the coefficient of variation (CV) of a random variable that follows a symmetric distribution in the real line. Specifically, we estimate this coefficient using the maximum-likelihood (ML) method. In addition, we provide asymptotic inference for this parameter, which allows us to contrast hypothesis and construct confidence intervals. Furthermore, we produce influence diagnostics to evaluate the sensitivity of the ML estimate of this coefficient when atypical data are present. Moreover, we illustrate the obtained results by using financial real data. Finally, we carry out a simulation study to detect the potential influence of atypical observations on the ML estimator of the CV of a symmetric distribution. The illustration and simulation demonstrate the robustness of the ML estimation of this coefficient.     

The International Evolution of Italian and Chinese Districts: What Role for Lead Firms?

The paper stresses how Italian and Chinese small and medium enterprises (SMEs) internationalize according to different approaches. Indeed, in the former case firms may move abroad because they follow the so-called lead firms that act as a meta-national body, whereas in the latter case SMEs are committed to exports and/or FDIs through a thick and wide web of ethnic networks. These are based on factors such as embeddedness and social capital as well as other cultural determinants.     

Probability inequalities for continuous unimodal random variables with finite support

A paramecer-free Bernstein-type upper bound is derived for the probability that the sum S of n i.i.d, unimodal random variables with finite support, X₁ ,X₂,…,X_n, exceeds its mean E(S) by the positive value nt. The bound for P{S - nμ ≥ nt} depends on the range of the summands, the sample size n, the positive number t, and the type of unimodality assumed for X_i. A two-sided Gauss-type probability inequality for sums of strongly unimodal random variables is also given. The new bounds are contrasted to Hoeffding's inequality for bounded random variables and to the Bienayme-Chebyshev inequality. Finally, the new inequalities are applied to a classic probability inequality example first published by Savage (1961).     

Designing a “Think-Along Dwelling” for People With Dementia: A Co-Creation Project Between Health Care and the Building Services Sector

Ageing-in-place is the preferred way of living for older individuals in an ageing society. It can be facilitated through architectural and technological solutions in the home environment. Dementia poses additional challenges when designing, constructing, or retrofitting housing facilities that support ageing-in-place. Older adults with dementia and their partners ask for living environments that support independence, compensate for declining and vitality, and lower the burden of family care. This study reports the design process of a demonstration home for people with dementia through performing a literature review and focus group sessions. This design incorporates modifications in terms of architecture, interior design, the indoor environment, and technological solutions. Current design guidelines are frequently based on small-scale studies, and, therefore, more systematic field research should be performed to provide further evidence for the efficacy of solutions. The dwellings of people with dementia are used to investigate the many aspects of supportive living environments for older adults with dementia and as educational and training settings for professionals from the fields of nursing, construction, and building services engineering.     

Modeling delay in diagnosis of NF: under reportincg,incidence and prevalence estimates

In this paper, we analyze data from the Italian National Register of Rare Diseases (NRRD) focusing, in particular, on the geo-temporal distribution of patients affected by neurofibromatosis type 1 (NF1, ICD9CM code 237.71). The aim is at deriving a corrected measure of incidence for the period 2007–2009 using a single source, and to provide NF1 prevalence estimates for the period 2001–2006 through the use of capture–recapture methods over two sources. In the first case, a reverse hazard estimator for the delay in diagnosis of NF1 is used to estimate the probability that a generic unit belonging to the population of interest has been registered by the archive of reference. For the second purpose, two-source capture–recapture methods have been used to estimate the number of NF1 prevalent units in Italy for the period 2001–2006, matching information provided by the NRRD and the national register of hospital discharge, Scheda di Dimissione Ospedaliera (in the following SDO), archives.